[FFmpeg-devel] Re: [PATCH 0/7] matroska: support of timecode

[Dave Rice via ffmpeg-devel <ffmpeg-devel@ffmpeg.org>]

> On Sep 9, 2025, at 8:37 AM, Jerome Martinez via ffmpeg-devel <ffmpeg-devel@ffmpeg.org> wrote:
> 
> This patch series adds support of timecode in Matroska files, specs are at:
> https://github.com/ietf-wg-cellar/matroska-specification/blob/master/cellar-codec/block_additional_mappings/smpte-st12-1-timecode.md
> 
> As ST 12-1 support was already partially implemented, using 32 bits rather than 64 bits (binary groups were not stored), the ST 12-1 side data format was changed from 32 bits to 64 bits rather than introducing another ST  12-1 side data internal "format", this may break code using it but not in the main repo (I found nothing using it in the main repo), is it fine?
> If we keep internally 32 bits for ST 12-1, any non zero timecode binary group stored in MKVs would be trashed.
> 
> The first 3 patches refactor BlockAddition support in MKV encoder in order to have it more versatile
> The 2 next patches adapt decklink_dec for writing 64 bit timecodes in side data rather than 32 bit timecodes
> The last 2 patches adds timecode support in Matroska

Tested with a decklink input with `./ffmpeg -y -f decklink -timecode_format rp188any -audio_input embedded -video_input sdi -format_code ntsc -i "UltraStudio 4K Mini" -c copy -t 1 output6.mkv`. Output includes the timecode values stored in Block Additionals. This is great!

Also noting that the timecode side data only seems to travel with streamcopy. If I switch the `-c ffv1` the side data is lost.

Kind Regards,
Dave

_______________________________________________
ffmpeg-devel mailing list -- ffmpeg-devel@ffmpeg.org
To unsubscribe send an email to ffmpeg-devel-leave@ffmpeg.org